Skip to content

Benchmark

Dubesor Censor

Scores are percentages.

Models measured
219
Scores recorded
261
Sources
1
IQ weight
0.24

Score distribution

52.5%
Mean
52.5%
Median
19.9%
Std dev
27.8%
10th percentile
80.2%
90th percentile
2.7%
Min
2.7%95.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.24 for chat models, measured over 219 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1MythoMax L2 13b· Dubesor LLM Benchmark95.5%
2DeepSeek R1 Zero· Dubesor LLM Benchmark93.2%
3WizardLM 2 8x22B· Dubesor LLM Benchmark91.9%
4Grok 4· Dubesor LLM Benchmark91.8%
5Grok 4.20reasoning· Dubesor LLM Benchmark91.8%
6Command R+· Dubesor LLM Benchmark90%
7Mistral Medium· Dubesor LLM Benchmark90%
8Grok 4.20non-reasoning· Dubesor LLM Benchmark89.3%
9Mistral Large2402· Dubesor LLM Benchmark87.7%
10Llama 3.1 70B Instruct· Dubesor LLM Benchmark86.9%
11Claude Sonnet 4.6· Dubesor LLM Benchmark86.4%
12GPT-4 Turbo· Dubesor LLM Benchmark86.4%
13Jamba 1.5 Large· Dubesor LLM Benchmark86.4%
14Llama 3.1 Nemotron Ultra 253B v1· Dubesor LLM Benchmark86.4%
15Magistral Medium2506· Dubesor LLM Benchmark86.4%
16Mistral Medium 3· Dubesor LLM Benchmark86.4%
17Yi Lightning· Dubesor LLM Benchmark85%
18Grok 2· Dubesor LLM Benchmark83.8%
19Mistral Large 2· Dubesor LLM Benchmark83.8%
20GPT-4o mini· Dubesor LLM Benchmark83%
21Llama 3.2 90B Vision Instruct· Dubesor LLM Benchmark82.4%
22GPT 4o· Dubesor LLM Benchmark81.9%
23Grok Code Fast 1· Dubesor LLM Benchmark81.2%
24OpenHermes 2.5 Mistral 7B· Dubesor LLM Benchmark81.2%
25Llama 3.1 405B Instruct· Dubesor LLM Benchmark81.1%
26Command A 03 2025· Dubesor LLM Benchmark80.2%
27Llama 3.1 Nemotron 70B Instruct· Dubesor LLM Benchmark80.2%
28jamba-1.6-mini· Dubesor LLM Benchmark79.6%
29Devstral 2 123b Instruct 2512· Dubesor LLM Benchmark78.2%
30Gemini 3 Flash· Dubesor LLM Benchmark78.2%
31Gemini 3.1 Flash Lite· Dubesor LLM Benchmark78.2%
32Grok 2 mini· Dubesor LLM Benchmark78.2%
33Grok 3· Dubesor LLM Benchmark78.2%
34Llama 3.3 Nemotron Super 49B v1· Dubesor LLM Benchmark78.2%
35Llama 4 Scout· Dubesor LLM Benchmark78.2%
36Magistral Small 2509· Dubesor LLM Benchmark78.2%
37Mistral Small 3.1 24b Instruct 2503· Dubesor LLM Benchmark78.2%
38DeepSeek V3· Dubesor LLM Benchmark76.7%
39DeepSeek V3.1· Dubesor LLM Benchmark76.7%
40Mixtral 8x7B Instruct v0.1· Dubesor LLM Benchmark76.4%
41yi-large· Dubesor LLM Benchmark76.4%
42Ministral 8B· Dubesor LLM Benchmark75.7%
43Gemma 4 26B A4B· Dubesor LLM Benchmark74.9%
44Llama 3.3 Nemotron Super 49B v1thinking· Dubesor LLM Benchmark74.9%
45Mistral Small 3.2 24B Instruct· Dubesor LLM Benchmark74.8%
46Llama 3.2 11B Vision Instruct· Dubesor LLM Benchmark74.6%
47Jamba 1.5 Mini· Dubesor LLM Benchmark74.4%
48DeepSeek V2.5· Dubesor LLM Benchmark74%
49Llama 3.3 70B Instruct· Dubesor LLM Benchmark73.3%
50Gemini 3 Flashthinking· Dubesor LLM Benchmark73.1%

Showing 1–50 of 261 results.